Wstepna analiza wykazala wiele niescislosci w podanym zestawie danych. Pierwszym problemem jest nie chronologiczne ustawienie danych przez co nie mozna jesdnoznacznie stwierdzic roku w którym wykonane byly pomiary. Takze dane pomiary z danego miesiaca sa pomieszane i nie sa pogrupowane razem. Zmienna totaln reprezentujaca ilosc sledzi jest podana jako number, a nie jako integer. W zestawie danych pojawia sie takze sporo wartosci NA jest ich: 11056. W tym przypadku przy tak duzej ilosci niepelnych danych nie mozna bylo ich poninac. Pierwszym podejciem bylo uzupelnienie dancyh wartosciami srednimi z danego misiaca niestety nie powtórna analiza wykazala, ze nie jest to odpowiednie podejscie. Po kolejnym przeanalizowaniu danych zawuwazona zaostala zaleznosc pomiedzy danymi oraz to, ze sie powtarzaja w grupach. Wiec dane puste zostaly zastapione takimi samymi wartosciami jak ich sasiedzi, którzy posiadaja takie same pozostaae wartossci. Dzieki temu dane sa spójne. Po wykonaniu histogramu zmiennej length okazalo sie, ze prezentuje on rozklad normalny. Nastepnie dane zostaly znormalizowane do przedzialu od 0 do 1 tak aby uniknac faworyzowania zmiennych. Znormalizowane zostaly tylko dane treningowe testowe pozostaly niezmienione. Jak widac na modelach jak i na macierzy korelacji najwiekszy wplyw na dlugosc sledzia ma temperatura przy powierzchni wody, a takze dostepnosc planktonu. Do regresji zostaly uzyte 2 metody Random Forest oraz Stochastic Gradient Boosting. Lepiej wypadla metoda Random forest choc nie pokazala, najwiekszej zaleznosci zmiennej length od temperatury przy powierzchni wody. Wydaje mi sie, ze duzym problememjest slaby opis danych, który jest niejednoznaczny.
Najwazeniejsze informacje na temat zbioru sledzi po usnieciu wartosci pustych oraz wstepnym przetworzeniu danych. Ponizej przedstawiona jest legenda, która przedstawia co oznaczaja poszczególne zmienne.
## X length cfin1 cfin2
## Min. : 0 Min. :19.0 Min. : 0.0000 Min. : 0.0000
## 1st Qu.:13145 1st Qu.:24.0 1st Qu.: 0.0000 1st Qu.: 0.2778
## Median :26291 Median :25.5 Median : 0.1111 Median : 0.7012
## Mean :26291 Mean :25.3 Mean : 0.4462 Mean : 2.0261
## 3rd Qu.:39436 3rd Qu.:26.5 3rd Qu.: 0.3333 3rd Qu.: 1.7936
## Max. :52581 Max. :32.5 Max. :37.6667 Max. :19.3958
## chel1 chel2 lcop1 lcop2
## Min. : 0.000 Min. : 5.238 Min. : 0.3074 Min. : 7.849
## 1st Qu.: 2.469 1st Qu.:13.427 1st Qu.: 2.5479 1st Qu.:17.808
## Median : 5.750 Median :21.435 Median : 7.0000 Median :24.859
## Mean :10.003 Mean :21.218 Mean : 12.8053 Mean :28.423
## 3rd Qu.:11.500 3rd Qu.:27.193 3rd Qu.: 21.2315 3rd Qu.:37.232
## Max. :75.000 Max. :57.706 Max. :115.5833 Max. :68.736
## fbar recr cumf totaln
## Min. :0.0680 Min. : 140515 Min. :0.06833 Min. : 144137
## 1st Qu.:0.2270 1st Qu.: 360061 1st Qu.:0.14809 1st Qu.: 306068
## Median :0.3320 Median : 421391 Median :0.23191 Median : 539558
## Mean :0.3304 Mean : 520367 Mean :0.22981 Mean : 514973
## 3rd Qu.:0.4560 3rd Qu.: 724151 3rd Qu.:0.29803 3rd Qu.: 730351
## Max. :0.8490 Max. :1565890 Max. :0.39801 Max. :1015595
## sst sal xmonth nao
## Min. :12.77 Min. :35.40 Min. : 1.000 Min. :-4.89000
## 1st Qu.:13.60 1st Qu.:35.51 1st Qu.: 5.000 1st Qu.:-1.89000
## Median :13.86 Median :35.51 Median : 8.000 Median : 0.20000
## Mean :13.88 Mean :35.51 Mean : 7.258 Mean :-0.09236
## 3rd Qu.:14.16 3rd Qu.:35.52 3rd Qu.: 9.000 3rd Qu.: 1.63000
## Max. :14.73 Max. :35.61 Max. :12.000 Max. : 5.08000
## [1] 52582
Jezeli zalozymy, ze dane sa poukladane chronologicznie to wykres zmiany dlugosci w czasie bedzie wygladal nastepujaco.
library(plotly)
##
## Attaching package: 'plotly'
## The following object is masked from 'package:ggplot2':
##
## last_plot
## The following objects are masked from 'package:plyr':
##
## arrange, mutate, rename, summarise
## The following object is masked from 'package:stats':
##
## filter
## The following object is masked from 'package:graphics':
##
## layout
library(ggplot2)
p<-ggplot(sledzie, aes(x = X, y=length)) + geom_point() + geom_smooth(method="auto", se=TRUE, color="red")
ggplotly(p)
## `geom_smooth()` using method = 'gam'
Jezeli natomiast dane nie sa posortowane chronologicznie mozna je pogrupowac po wartosci zmiennej recr która mówi o rocznym polowie a wiec w sposób nie bezposredni definiuje nam poszczególne lata.
Jak widac trend jest dosyc podobny w obu przypadkach.
Aby miec wiekszy poglad na te dane zobaczmy jak ksztaltowaly sie wartosci zmiennych w poszczególnych latach(srednie wartosci zmiennych dla poszczególnych wartosci zmiennej recr):
library(plyr)
library(dplyr)
library(dplyr)
library(printr)
var<-sledzie%>%group_by(recr)%>%summarize(mean_cfin1=mean(cfin1),mean_cfin2=mean(cfin2),
mean_chel1=mean(chel1),mean_chel2=mean(chel2),
mean_lcop1=mean(lcop1),mean_lcop2=mean(lcop2),
mean_fbar=mean(fbar),mean_cum=mean(cumf),
mean_totaln=mean(totaln),mean_sst=mean(sst),
mean_sal=mean(sal),mean_nao=mean(nao))
head(var,length(var$recr))
| recr | mean_cfin1 | mean_cfin2 | mean_chel1 | mean_chel2 | mean_lcop1 | mean_lcop2 | mean_fbar | mean_cum | mean_totaln | mean_sst | mean_sal | mean_nao |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 140515 | 0.0002490 | 1.7928143 | 3.711355 | 31.426140 | 3.914250 | 36.571965 | 0.3370000 | 0.3461349 | 182192.7 | 14.13613 | 35.51955 | 0.7600000 |
| 148045 | 0.0000000 | 0.7272441 | 2.594878 | 33.115302 | 5.350880 | 37.210289 | 0.5470000 | 0.2714717 | 179955.4 | 13.39543 | 35.49992 | -2.1400000 |
| 163005 | 0.2181000 | 0.5777800 | 3.196834 | 20.158120 | 4.261900 | 39.902140 | 0.3900000 | 0.2571368 | 147331.7 | 13.90653 | 35.51142 | 2.5200000 |
| 168531 | 0.2053600 | 19.3958300 | 11.245540 | 32.649400 | 11.776790 | 65.455210 | 0.4670000 | 0.3854825 | 201854.9 | 13.19707 | 35.47843 | -2.2500000 |
| 186562 | 0.1005300 | 0.0000000 | 5.095260 | 15.318104 | 5.717370 | 16.842920 | 0.4250000 | 0.2176548 | 158165.2 | 14.05720 | 35.51374 | 1.3700000 |
| 204165 | 0.0875631 | 0.0256400 | 7.887500 | 33.508820 | 9.018750 | 39.183820 | 0.2920000 | 0.2147636 | 160242.8 | 13.56747 | 35.48954 | 1.2300000 |
| 208551 | 0.1838400 | 2.0001612 | 1.924070 | 17.889270 | 2.187880 | 22.098554 | 0.3370000 | 0.2728057 | 539558.4 | 14.42667 | 35.51339 | 3.9600000 |
| 227463 | 0.1000000 | 0.3714300 | 3.880445 | 25.840100 | 5.187500 | 28.050717 | 0.2680000 | 0.1883533 | 172576.8 | 13.87040 | 35.50400 | 1.6300000 |
| 247178 | 0.0400000 | 0.8048900 | 6.527310 | 18.628027 | 6.746410 | 22.970922 | 0.6880000 | 0.3980148 | 307276.5 | 14.72520 | 35.51633 | -0.1700000 |
| 254830 | 0.0000000 | 0.9224700 | 75.000000 | 30.408100 | 74.969812 | 38.220573 | 0.5670000 | 0.2980328 | 568477.6 | 13.59640 | 35.46781 | -4.8900000 |
| 282152 | 0.0370400 | 0.0000000 | 0.228700 | 15.832320 | 0.317545 | 16.139679 | 0.8490000 | 0.3630819 | 144136.7 | 13.37547 | 35.50901 | 0.1700000 |
| 282493 | 0.4178600 | 0.5877900 | 2.704400 | 10.947900 | 3.499760 | 15.003140 | 0.5000000 | 0.2866648 | 306160.3 | 13.60000 | 35.52906 | -0.9600000 |
| 327066 | 0.0294985 | 0.2504959 | 8.662444 | 27.192860 | 8.984390 | 29.605064 | 0.5210000 | 0.2767387 | 194320.1 | 13.06587 | 35.49925 | 0.3400000 |
| 351797 | 0.7459915 | 0.6335547 | 14.679700 | 28.045260 | 16.365590 | 35.152996 | 0.4650000 | 0.2811635 | 375469.5 | 14.55720 | 35.61034 | 3.9600000 |
| 355107 | 0.0000000 | 0.0000000 | 4.811380 | 12.728350 | 5.165740 | 15.893640 | 0.0830000 | 0.0736151 | 595514.4 | 14.44133 | 35.45592 | 2.0500000 |
| 359652 | 0.1333300 | 1.5804600 | 1.879170 | 14.521228 | 2.234720 | 20.093760 | 0.3990000 | 0.2243824 | 555586.3 | 14.02253 | 35.51105 | 3.2800000 |
| 360061 | 0.0000431 | 0.1173600 | 12.151920 | 39.568090 | 12.495880 | 41.628954 | 0.1380000 | 0.2665911 | 289260.6 | 14.44160 | 35.39803 | 0.2000000 |
| 364794 | 3.0000000 | 3.0448993 | 4.000000 | 16.128790 | 7.000000 | 23.366520 | 0.2000000 | 0.1095853 | 655248.7 | 14.39387 | 35.51206 | -0.3700000 |
| 370511 | 0.0100000 | 0.2671300 | 6.571670 | 37.638010 | 7.071670 | 39.981285 | 0.6630000 | 0.3502113 | 279064.5 | 13.85867 | 35.53935 | -3.7800000 |
| 373947 | 0.1000000 | 1.0912700 | 2.266790 | 24.837650 | 2.503010 | 31.027950 | 0.0740000 | 0.1033722 | 396283.7 | 14.65293 | 35.55342 | 1.0918788 |
| 392084 | 0.9418352 | 0.2960097 | 6.138738 | 21.668940 | 8.982271 | 24.858672 | 0.1580000 | 0.1100757 | 766077.6 | 14.06922 | 35.51526 | -1.5400000 |
| 405494 | 0.0000000 | 0.1870000 | 0.000000 | 11.116160 | 1.750000 | 20.965330 | 0.1410000 | 0.0755325 | 1015594.9 | 13.28067 | 35.51322 | -2.3800000 |
| 421391 | 0.0000000 | 0.7011800 | 11.499943 | 5.683691 | 22.992102 | 9.191343 | 0.2270000 | 0.1480941 | 730351.2 | 13.63160 | 35.50835 | -2.8600000 |
| 423281 | 0.0684433 | 1.4159623 | 7.751343 | 9.417813 | 14.490526 | 14.156656 | 0.1360000 | 0.1094430 | 904060.3 | 13.75026 | 35.50865 | -0.4553333 |
| 441827 | 0.3595984 | 5.3640200 | 4.326740 | 27.112108 | 5.071755 | 36.626280 | 0.4340000 | 0.3726272 | 191976.2 | 14.47771 | 35.50777 | -1.9000000 |
| 459347 | 0.0933000 | 6.5288400 | 4.315840 | 28.070770 | 4.469810 | 42.959895 | 0.2310000 | 0.2590166 | 264308.2 | 14.57347 | 35.50983 | 1.7000000 |
| 465638 | 0.1894000 | 0.8568400 | 0.603080 | 9.432080 | 0.828030 | 10.761666 | 0.5710000 | 0.3500081 | 383913.5 | 13.86246 | 35.51779 | 0.5600000 |
| 469158 | 4.8333300 | 4.2111900 | 36.333330 | 57.626808 | 41.166670 | 68.598205 | 0.4560000 | 0.2773366 | 413634.1 | 13.73827 | 35.46280 | -1.0400000 |
| 473462 | 0.0545500 | 0.2313400 | 1.488682 | 11.670930 | 1.835150 | 17.807530 | 0.5910000 | 0.3758273 | 306067.6 | 14.35618 | 35.52329 | 0.7200000 |
| 474983 | 3.1446200 | 5.8014500 | 4.261480 | 18.438390 | 7.671480 | 27.470012 | 0.2540000 | 0.2073709 | 534157.2 | 13.89253 | 35.50455 | -0.7500000 |
| 482831 | 0.0277800 | 0.2778500 | 2.471847 | 21.439224 | 2.547870 | 26.363420 | 0.3560000 | 0.3059879 | 267380.8 | 14.30693 | 35.51234 | 2.8000000 |
| 503264 | 0.0860538 | 0.5722722 | 1.366600 | 5.237640 | 1.531640 | 7.969152 | 0.5030000 | 0.3069927 | 329159.4 | 14.65493 | 35.52548 | 5.0800000 |
| 574641 | 0.2400000 | 4.9180700 | 6.082860 | 13.589210 | 7.122860 | 22.667706 | 0.3910000 | 0.2315489 | 514114.5 | 13.51507 | 35.50230 | 0.7200000 |
| 640184 | 0.2344300 | 1.5396000 | 1.372060 | 13.426620 | 1.853220 | 20.296208 | 0.4030000 | 0.2585906 | 519512.4 | 14.15572 | 35.43245 | 1.6000000 |
| 650742 | 1.2133300 | 4.5582500 | 19.154750 | 26.803755 | 21.231470 | 45.677052 | 0.2390000 | 0.2223979 | 676596.4 | 13.55987 | 35.52449 | -0.6300000 |
| 664944 | 0.6171000 | 13.1433800 | 3.590480 | 33.909170 | 5.212690 | 64.823460 | 0.3760000 | 0.2327722 | 542230.0 | 13.98507 | 35.50663 | 3.4200000 |
| 717939 | 0.1415800 | 0.3020300 | 2.030660 | 20.123260 | 2.241147 | 24.081406 | 0.4220000 | 0.2532627 | 460804.1 | 13.66578 | 35.51403 | 3.0300000 |
| 724151 | 1.0250800 | 3.6631900 | 6.421270 | 25.508235 | 10.928570 | 37.392010 | 0.4850000 | 0.3838187 | 457143.9 | 13.71160 | 35.51169 | 2.0500000 |
| 741245 | 0.1111100 | 1.5690553 | 32.000000 | 26.310961 | 33.333330 | 36.193003 | 0.2440000 | 0.1640387 | 763082.9 | 13.48293 | 35.52719 | -2.8800000 |
| 766083 | 0.0357521 | 1.0734500 | 0.766420 | 10.109630 | 1.342720 | 14.554657 | 0.3180000 | 0.2324083 | 559434.9 | 13.79867 | 35.51146 | 2.6700000 |
| 774993 | 0.8412095 | 0.0007815 | 22.659978 | 15.038492 | 29.069180 | 17.693025 | 0.3670000 | 0.2035341 | 826464.9 | 13.69493 | 35.54564 | -1.6900000 |
| 783337 | 2.1433300 | 4.4588200 | 6.386670 | 26.171870 | 9.010000 | 32.190900 | 0.3270000 | 0.3096315 | 492519.0 | 13.98122 | 35.61240 | 0.8000000 |
| 824154 | 0.0000000 | 0.2450777 | 30.833330 | 28.213380 | 31.500000 | 31.376650 | 0.5230000 | 0.2951823 | 389403.1 | 13.69187 | 35.49790 | -1.8900000 |
| 833003 | 0.2000000 | 0.0131200 | 2.418780 | 17.208940 | 2.810200 | 22.684790 | 0.0680000 | 0.0683259 | 363016.8 | 14.72947 | 35.58002 | 2.0500000 |
| 837339 | 0.8160084 | 0.3669823 | 6.051593 | 15.537214 | 9.683288 | 22.827520 | 0.0980000 | 0.0779181 | 631877.9 | 14.21173 | 35.51007 | 1.8000000 |
| 907207 | 0.3333300 | 0.1835300 | 9.719020 | 17.538910 | 27.333330 | 25.373964 | 0.3320000 | 0.2319097 | 597698.7 | 13.63173 | 35.52012 | 1.2800000 |
| 958184 | 0.8968600 | 6.0374000 | 3.022680 | 16.159164 | 4.424930 | 23.850791 | 0.3300000 | 0.2063772 | 482348.4 | 13.64259 | 35.51181 | 1.0300000 |
| 1079510 | 0.1416700 | 0.0845326 | 7.646430 | 29.719699 | 13.088100 | 30.834044 | 0.1300000 | 0.1771845 | 303522.2 | 14.51280 | 35.59639 | -0.0700000 |
| 1193220 | 0.3183397 | 2.5139182 | 6.943050 | 25.269898 | 9.022470 | 28.709176 | 0.2920000 | 0.2065740 | 589460.5 | 12.77135 | 35.52640 | 0.5000000 |
| 1322000 | 0.0000000 | 0.0100000 | 1.021430 | 26.006170 | 1.064290 | 34.145600 | 0.1000000 | 0.0922202 | 648314.9 | 14.55560 | 35.53620 | 2.0500000 |
| 1380210 | 0.1666700 | 0.5566776 | 5.750000 | 36.599190 | 5.945652 | 45.358867 | 0.1998266 | 0.1069226 | 774752.4 | 13.06453 | 35.51174 | -3.6000000 |
| 1565890 | 37.6666700 | 10.1696200 | 64.750000 | 43.644700 | 115.583330 | 59.085240 | 0.1250000 | 0.0958601 | 727441.4 | 13.61893 | 35.53495 | -1.0200000 |
| # Korelacj | a atrybutów |
| length | cfin1 | cfin2 | chel1 | chel2 | lcop1 | lcop2 | fbar | recr | cumf | totaln | sst | sal | nao | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| length | 1.00 | 0.08 | 0.10 | 0.22 | -0.01 | 0.24 | 0.05 | 0.25 | -0.01 | 0.01 | 0.10 | -0.45 | 0.03 | -0.26 |
| cfin1 | 0.08 | 1.00 | 0.15 | 0.09 | 0.20 | 0.12 | 0.21 | -0.06 | 0.12 | -0.05 | 0.13 | 0.01 | 0.13 | 0.01 |
| cfin2 | 0.10 | 0.15 | 1.00 | 0.00 | 0.31 | -0.04 | 0.65 | 0.15 | -0.10 | 0.34 | -0.22 | -0.24 | -0.08 | -0.01 |
| chel1 | 0.22 | 0.09 | 0.00 | 1.00 | 0.29 | 0.96 | 0.25 | 0.16 | -0.05 | 0.07 | 0.17 | -0.22 | -0.15 | -0.51 |
| chel2 | -0.01 | 0.20 | 0.31 | 0.29 | 1.00 | 0.18 | 0.88 | 0.03 | 0.00 | 0.26 | -0.38 | 0.01 | -0.22 | -0.06 |
| lcop1 | 0.24 | 0.12 | -0.04 | 0.96 | 0.18 | 1.00 | 0.15 | 0.10 | 0.00 | -0.01 | 0.27 | -0.26 | -0.10 | -0.55 |
| lcop2 | 0.05 | 0.21 | 0.65 | 0.25 | 0.88 | 0.15 | 1.00 | 0.05 | 0.00 | 0.29 | -0.30 | -0.12 | -0.18 | -0.04 |
| fbar | 0.25 | -0.06 | 0.15 | 0.16 | 0.03 | 0.10 | 0.05 | 1.00 | -0.24 | 0.82 | -0.51 | -0.18 | 0.04 | 0.07 |
| recr | -0.01 | 0.12 | -0.10 | -0.05 | 0.00 | 0.00 | 0.00 | -0.24 | 1.00 | -0.26 | 0.37 | -0.20 | 0.28 | 0.09 |
| cumf | 0.01 | -0.05 | 0.34 | 0.07 | 0.26 | -0.01 | 0.29 | 0.82 | -0.26 | 1.00 | -0.71 | 0.03 | -0.10 | 0.23 |
| totaln | 0.10 | 0.13 | -0.22 | 0.17 | -0.38 | 0.27 | -0.30 | -0.51 | 0.37 | -0.71 | 1.00 | -0.29 | 0.15 | -0.39 |
| sst | -0.45 | 0.01 | -0.24 | -0.22 | 0.01 | -0.26 | -0.12 | -0.18 | -0.20 | 0.03 | -0.29 | 1.00 | 0.01 | 0.51 |
| sal | 0.03 | 0.13 | -0.08 | -0.15 | -0.22 | -0.10 | -0.18 | 0.04 | 0.28 | -0.10 | 0.15 | 0.01 | 1.00 | 0.13 |
| nao | -0.26 | 0.01 | -0.01 | -0.51 | -0.06 | -0.55 | -0.04 | 0.07 | 0.09 | 0.23 | -0.39 | 0.51 | 0.13 | 1.00 |
Jak widac mam trzy pary skorelowanych dodatnich ze saba zmiennych :
0.960.880.82Oraz jedna pare zmiennych skorelowanych ujemnie:
-0.71Dla przykladu jak widac zmienne lcop1,chal1 w polaczeniu ze zmienna length maja bardzo podobna wartosc:
Usuniecie silnie skorlowawnych danch:
Wartosc RMSE dla random forrest o paramterach number = 2 i repeats = 5:
## [1] 1.639593
Wynik zbioru trningowego:
| mtry | RMSE | Rsquared | RMSESD | RsquaredSD |
|---|---|---|---|---|
| 2 | 1.186616 | 0.4836092 | 0.0034619 | 0.0021886 |
| 6 | 1.187033 | 0.4832753 | 0.0035525 | 0.0021495 |
| 10 | 1.187323 | 0.4830191 | 0.0033672 | 0.0019964 |
Wartosc RMSE dla Stochastic Gradient Boosting:
## [1] 1.828024
Wynik zbioru trningowego:
| shrinkage | interaction.depth | n.minobsinnode | n.trees | RMSE | Rsquared | RMSESD | RsquaredSD | |
|---|---|---|---|---|---|---|---|---|
| 1 | 0.1 | 1 | 10 | 50 | 1.295578 | 0.4004936 | 0.0111707 | 0.0102719 |
| 4 | 0.1 | 2 | 10 | 50 | 1.234282 | 0.4469241 | 0.0116917 | 0.0102356 |
| 7 | 0.1 | 3 | 10 | 50 | 1.210377 | 0.4654263 | 0.0125273 | 0.0106578 |
| 2 | 0.1 | 1 | 10 | 100 | 1.257651 | 0.4249372 | 0.0116365 | 0.0100652 |
| 5 | 0.1 | 2 | 10 | 100 | 1.206139 | 0.4680349 | 0.0128815 | 0.0105481 |
| 8 | 0.1 | 3 | 10 | 100 | 1.193233 | 0.4782504 | 0.0136845 | 0.0110176 |
| 3 | 0.1 | 1 | 10 | 150 | 1.239350 | 0.4401986 | 0.0121557 | 0.0103026 |
| 6 | 0.1 | 2 | 10 | 150 | 1.196407 | 0.4756238 | 0.0137473 | 0.0111532 |
| 9 | 0.1 | 3 | 10 | 150 | 1.188967 | 0.4817249 | 0.0140189 | 0.0110185 |
varImp(fit$finalModel)
| Overall | |
|---|---|
| cfin1 | 3.111408 |
| cfin2 | 5.747870 |
| lcop1 | 6.637928 |
| lcop2 | 7.127210 |
| fbar | 4.080895 |
| recr | 4.104960 |
| totaln | 6.427989 |
| sst | 7.194826 |
| sal | 3.943398 |
| nao | 1.696503 |
varImp(gbmFit$finalModel)
| Overall | |
|---|---|
| cfin1 | 3909.3983 |
| cfin2 | 3825.5970 |
| lcop1 | 9907.7995 |
| lcop2 | 7439.6102 |
| fbar | 6076.4766 |
| recr | 19842.7477 |
| totaln | 9304.6444 |
| sst | 75322.5532 |
| sal | 976.1131 |
| nao | 3069.9548 |